This "Best Paper Award" assesses the top scholarly security research from the previous year. The focus of this competition is on research that has a practical impact.  With eligibility limited to already published papers or camera-ready papers, CSAW has a reputation for drawing some of the best security research worldwide. 

With the rise of large language models (LLMs), AI systems are now capable of identifying software vulnerabilities and generating exploit code—skills that align closely with the goals of Capture the Flag (CTF) competitions, where participants solve security challenges to uncover hidden “flags.”

In this competition, your task is to build your own agentic AI to solve CTF challenges autonomously—that is, create or extend an AI agent powered by LLMs to analyze and exploit challenges without human intervention. You may either bring your own agent framework or enhance a provided baseline agent, with technical support available. Your agent can leverage LLMs, either API-based models such as GPT, Claude and Gemini, or open-source models deployed locally to navigate common CTF categories such as crypto, forensics, pwn, reverse, web, and misc.

A successful submission includes the full logs of prompts and model responses, along with a brief write-up describing your prompting strategies, agent enhancements, and system design choices.

Participating teams will be tasked with leveraging AI tools, such as LLMs, to insert and exploit hardware vulnerabilities and Trojans for various open-source hardware designs. These can include cryptographic accelerators, processors, communication IPs, etc.

Each month leading up to the in-person final at CSAW, a new challenge (or challenges) will be issued. Each of these challenges will remain available for the duration of the competition until the finalist teams are selected so new teams can join at any time. At the conclusion of each month’s challenges, a winning team will be selected and awarded a small hardware prize.

Potential challenges could include:

• Inserting a Trojan that can evade detection by state of the art security analysis tools and models. 
• Modifying security checks to allow a Trojan-infected hardware module to pass verification.
• Modifying a design to make it more vulnerable to potential side-channel attacks.

All challenges must be completed using AI tools and all information regarding those tools must be submitted, including complete and detailed logs of their use, i.e. if an LLM is being used, we require all conversations with the model be submitted.